The objective of this project is to gain a better understanding of SPEM development and control creating, testing and refining a computational modeling framework of SPEM control. The model we propose combines the useful properties of two different SPERM models: negative feedback and target estimation. The model creates a high-level central representation of target velocity. Simultaneously, the model behaves like feedback controller which depends upon visual feedback delay. We will perform parametric fits of SPEM behavioral data using a preliminary model. This operation will : 1) establish model parameter values and 2) show if revision to the preliminary model are required. The model will contain adaptive mechanisms which develop and maintain values of the model. We also propose and perform and experiment which will serve as a test of our model. The experiment will test the ability to track an invisible target whose velocity is determined while the eyes are stationary. We will use the results of the experiment refine our model.